Two primary characteristics of cloud computing are energy consumption and execution time optimization. The massive amount of data that needs to be processed grows as the number of data centers does. One of the most popular strategies for reducing energy waste while making the best use of available resources is the efficient scheduling of user processes carried out in the cloud. In this research, we have utilized a meta-heuristic with a sensitivity rate named (MHSC) to reduce energy consumption and execution time while maximizing throughput. Our method combined two meta-heuristic algorithms named Genetic and Bat and benefited from their advantages while omitting their shortcoming to find global optima. The proposed method intends to simultaneously optimize throughput, execution time, and energy usage. By combining special parameters, we reached the objective function. By weighing the jobs to cluster the workflow inputs, utilizing linear programming of hybrid bat and genetic algorithms to generate an inertia function, and accounting for a ratio function to control the number of iterations, MHSC algorithm reached a better optimum point. It allowed a combination of genetic and bat algorithms to discover the global optimum more quickly and escape from local optimal traps. By enhancing the echolocation parameter Ajk we were able to outperform the traditional bat algorithm. By Combining sensitivity rate, neighborhood type, and neighborhood address, the address of quality of the data link was predicted, and using the first threshold detector and sensitivity rate, the intelligent threshold was calculated. We carried out several experiments on a variety of workflows, from those with low dependence to those with significant dependency. The results showed that the MHSC outperforms its competitors as the number of nodes and the dependencies between them increase. Energy consumption and execution time improved by 6.4% and 8.1% respectively.

中文翻译：

---

MHSC：一种使用灵敏度率计算优化吞吐量和能量的元启发式方法


云计算的两个主要特征是能耗和执行时间优化。需要处理的大量数据随着数据中心数量的增加而增长。在充分利用可用资源的同时减少能源浪费的最流行策略之一是在云中执行的用户流程的高效调度。在这项研究中，我们利用了一种灵敏度为 （MHSC） 的元启发式方法来降低能耗和执行时间，同时最大限度地提高吞吐量。我们的方法结合了两种名为 Genetic 和 Bat 的元启发式算法，并受益于它们的优势，同时省略了它们的缺点来寻找全局最优值。所提出的方法旨在同时优化吞吐量、执行时间和能源使用。通过组合特殊参数，我们得到了目标函数。通过权衡作业以对工作流输入进行聚类，利用混合 bat 和遗传算法的线性编程来生成惯性函数，并考虑比率函数来控制迭代次数，MHSC 算法达到了更好的最佳点。它允许遗传算法和 bat 算法的组合更快地发现全局最优值并摆脱局部最优陷阱。通过增强 echolocation 参数 Ajk，我们能够超越传统的 bat 算法。通过结合灵敏度率、邻域类型和邻域地址，预测数据链路的质量地址，并使用第一阈值检测器和灵敏度计算智能阈值。我们对各种工作流程进行了多项实验，从低依赖性到显著依赖性。 结果表明，随着节点数量和节点之间的依赖关系的增加，MHSC 的表现优于其竞争对手。能耗和执行时间分别提高了 6.4% 和 8.1%。